Fundamentals of Stem Cells

Alt et al. Adipose-derived regenerative cells
1
Correspondence
Dr. Eckhard Alt, MD
2210 Chilton Road
Houston, TX 77019, USA
Phone: +1-504-988-3040
Cell: +1-832-853-3898
E-mail: ealt@tulane.edu
Fundamentals of stem cells: why and
how patients' own adult stem cells
are the next generation of medicine
Eckhard U. Alt1-4
, Christoph Schmitz5
, Xiaowen Bai1,3,6
1
Heart and Vascular Institute, Department of Medicine, Tulane University Health Science
Center, New Orleans, LA, USA
2
Sanford Health, University of South Dakota, Sioux Falls, SD, USA
3
University of Texas MD Anderson Cancer Center, Houston, TX, USA
4
Isar Klinikum Munich, Munich, Germany
5
Chair of Neuroanatomy, Institute of Anatomy, Faculty of Medicine, Ludwig-Maximilian
University Munich, Munich, Germany
6
Department of Cell Biology, Neurobiology & Anatomy, Medical College of Wisconsin,
Milwaukee WI, USA
Various tissue resident stem cells are receiving attention from basic scientists
and clinicians as they hold certain promise for regenerative medicine. This
paper is intended to clarify and facilitate the understanding, development and
adoption of regenerative medicine in general and specifically of therapies
based on unmodified, autologous adipose-derived regenerative cells (UA-
ADRCs). To this end, results of landmark experiments on stem cells and
stem cell therapy performed in the labs of the authors are summarized, the
most intriguing of which are the following: (i) vascular associated
mesenchymal stem cells (MSCs) can be isolated from different organs
(adipose tissue, heart, skin, bone marrow and skeletal muscle) and
differentiated into ectoderm, mesoderm and endoderm, providing significant
support for the hypothesis of the existence of a small, ubiquitously
distributed, universal vascular associated stem cell with full pluripotency; (ii)
the orientation and differentiation of MSCs are driven by signals of the
respective microenvironment; and (iii) these stem cells irrespective of the
tissue origin exhibit full pluripotent differentiation potential without any prior
genetic modification or the need for culturing. They can be obtained from a
small amount of adipose tissue when using the appropriate technology for
isolating the cells, and can be harvested from and re-applied to the same
patient at the point of care without the need for complicated processing,
manipulation, culturing, expensive equipment, or repeat interventions. These
findings demonstrate the potential of UA-ADRCs for triggering the
development of an entire new generation of medicine for the benefit of
patients and of healthcare systems.
Key words: adipose-derived; bone regeneration; cartilage regeneration; clinical
application; clinical studies; differentiation; hair loss; induced pluripotent stem
cells; maxillary sinus augmentation; osteoarthritis; pluripotency; regenerative
cells; regenerative medicine; progenitor cells; scar tissue; stem cells; wound
healing
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 April 2019 doi:10.20944/preprints201904.0200.v1
© 2019 by the author(s). Distributed under a Creative Commons CC BY license.

2
1. Introduction
Regenerative cell therapy, which refers to the therapeutic
application of stem cells to repair diseased or injured
tissue, has received increasing attention from basic
scientists, clinicians and the public. Stem cells hold
significant promise for tissue regeneration due to their
innate ability to provide a renewable supply of progenitor
cells that can form multiple cell types, whole tissue
structures, and even organs. Stem cells are present in the
human body at all stages of life from the earliest times of
an embryo through adulthood and senescence.
Embryonic stem cells (ESCs), which can be derived
from the blastocyst stage of mammalian embryos, exhibit
remarkable developmental plasticity that enables them to
differentiate into all types of cells present in the body
(Vazin and Freed, 2010). However, several major issues,
including ethical concerns, potential allogenic immune
rejection, and the risk of teratoma formation have limited
research and clinical application of therapies based on
ESCs (c.f., e.g., Fujikawa et al., 2005; Swijnenburg et al.,
2005; King and Perrin, 2014). Primarily due to these
issues a major interest now largely shifted to stem cells
derived from postnatal and adult tissues. Most
prominently, researchers have identified in the vascular
location of all organs and tissues a type of adult stem cell,
commonly referred to as mesenchymal stem cells (MSCs)
that exhibit significant potential for differentiation into a
number of cell types (c.f., e.g., da Silva Meirelles,
Chagastelles and Nardi, 2006; Bai et al., 2007; Ilmer et al.,
2014). Although these adult stem cells are essentially
ubiquitous, isolation from critical organs such as bone
marrow, skin, skeletal muscle, liver, heart or brain has
limited practicality. This is due to the fact that such
isolation carries a potential risk of damage to the donor
and often does not yield enough cells. As such, it is
typically required to expand cells in culture in order to
obtain the desired number of cells for the respective
therapeutic application. A highly attractive alternative
with great clinical significance is to isolate these adult
stem cells from adipose tissue (outlined in detail in
Section 4 of this article).
The different terms and definitions used in this article to
describe adult stem cells are summarized in Figure 1.
Ethical approvals are summarized at the end of the
manuscript.
FIGURE 1  Summary of the different terms and definitions used in this article to describe adult stem cells. MSCs, mesenchymal stem
cells or mesenchymal stromal cells.

3
2. The definition of the term stem cells
2.1. The current state of confusion about the term
stem cells
Approximately 50 years ago Friedenstein et al. (1968;
1974) described multipotent cells derived from bone
marrow stroma that were adherent to plastic and had the
ability to differentiate into cells other than hematopoietic
lineages. Since then, our knowledge about stem cells has
been steadily growing. However, the current definition of
the term stem cells as given by Dominici et al. (2006), that
describes MSCs as being adherent to plastic, expressing
the surface markers CD73, CD90 and CD105, and having
the ability to differentiate into osteoblasts, adipocytes and
chondrocytes, is not quite complete. This is partly due to
the fact that, for example, fibroblasts express these surface
markers as well, without having the ability to
transdifferentiate into other lineages or being MSCs (Alt
et al., 2011). The definition of the term stem cells
primarily by surface markers (c.f. Dominici et al., 2006) is
not sufficient, especially as we have learned in the
meantime that the true pluripotent stem cells do not yet
express CD73, CD90 and CD105 (outlined in detail in
Section 3 of this article). Rather, expression of cell surface
markers is a dynamic process. When cultured in fetal
bovine serum (FBS) or platelet lysate culture media,
MSCs can turn on new surface markers. Alternatively,
MSCs in culture can lose their surface marker expression,
such as for example the loss of the previously expressed
progenitor marker CD34 or the endothelial progenitor
marker CD31. In addition, the specific type of culture
media (FBS or defined serum free media [SFM]) and the
surface characteristics of a culture dish strongly impact on
visual appearance, growth rate and surface marker
expression of MSCs. For example, MSCs are adherent to
plastic and display a typical, spindle shaped appearance
when cultured in FBS (Fig. 2a,c). In contrast, they form
spheroids when cultured in SFM (Fig. 2b,d). Besides this,
the cellular growth rate determined by the G1, S and M
phases of the cell cycle considerably differ between MSCs
cultured in FBS (high growth rate) compared with MSCs
cultured in SFM (very low growth rate) (data not shown).
FIGURE 2  Change in the morphology of rat adipose-derived stem cells cultured first in fetal bovine serum (FBS) for 14 days (a), then
in serum free medium (SFM) for 24 hours (b), and then again in FBS for one week (c). Afterwards, then transfer to SFM again induced
a spheroid like appearance (d). The scale bar represents 100 µm.
One of the major current problems in understanding
stem cells is based on different and varying definitions of
the meaning of the term stem cells (e.g., Pittenger et al.,
1999; Phinney and Prockop, 2007; Bourin et al., 2013).
The way the term stem cells is often used by researchers
and the public confuses true stem cells with progenitor
cells. True stem cells are defined as being able to
differentiate into all three germ layers (e.g., Bai et al.,
2007; Ratajczak, 2015). In contrast, progenitor cells are
already on a pre-determined pathway to become a
differentiated cell and have lost their ability to decide
what they want to be ‘in life’. In other words, in contrast
to true stem cells, progenitor cells are typically determined
to differentiate and develop into a defined cell type and
primarily have lost their multipotency. For example, in
bone marrow, more than 99% of the cells are not true stem
cells, but primarily hematopoietic progenitor cells (e.g.,
Song, Prantl and Alt, 2010; Bruno et al., 2014).
Accordingly, progenitor cells have already started a
pathway of lineage-committed differentiation. In case of
bone marrow derived cells, these hematopoietic progenitor
cells (often incorrectly labeled as stem cells) started to
differentiate into future hematopoietic cells of the white,
red or platelet lineage. Pending on their progress in
maturation in this differentiation process, these cells are
no longer able to revert their pathway of differentiation.
At best, they are able to vary somewhat within the same
germ layer of differentiation, but typically stay within the
same lineage (Ricci-Vitiani et al., 2010; Wang et al.,
2010). Much of the confusion about the term stem cells

4
originates from this mix-up of the definition of true stem
cells with progenitor cells in the literature (c.f., e.g.,
Bourin et al., 2013; Caplan, 2017; Velten et al., 2017; see
also Alt, 2015).
2.2. The limited significance of induced
pluripotent stem cells for the practice of
medicine
Another problem is the incorrect public belief that
naturally no stem cells would exist that are able to
differentiate into all three lineages without being first
modified or genetically altered. The euphoria around the
so-called induced pluripotent stem cells (iPS cells) (that
resulted in the granting of the Nobel Prize in Physiology
or Medicine 2012) has led to a direction in the last couple
of years in research that will likely not significantly
change the practice of medicine. For more than ten years it
has been demonstrated many times that naturally there are
pluripotent stem cells which are able to differentiate into
all three germ layers without being first genetically
modified (c.f., e.g., Pittenger et al., 1999; Zuk et al., 2001;
Song, Prantl and Alt, 2011). Hence, it becomes clear that
the need for genetically manipulated iPS cells (in which
first an artificial [induced] overexpression of embryonic
genes like Oct4, Klf4, Sox2 and/or cMyc is necessary; c.f.,
e.g., Jullien et al., 2011; Heffernan, Sumer and Verma,
2011; Shi et al., 2017) is not required for the practice of
medicine. Rather, the universally distributed true stem
cells we refer to in this article are naturally able to
differentiate into all three germ layers and into any tissue
of the body under guidance of the local
microenvironment, without requiring prior overexpression
of embryonic genes or any prior artificial genetic
manipulation.
Another important aspect regarding iPS cells is that the
genetic manipulation of overexpression of embryonic
genes in these cells makes them less likely to enter the
practice of medicine due to the complexity of the
procedure. Besides this, there are concerns that iPS cells
can demonstrate features similar to cancer cells (c.f., e.g.,
Takahashi et al., 2007; Yu et al., 2007; Lee et al., 2013).
While the iPS cell technology will most likely not advance
to a stage where therapeutic transplants are deemed safe
(Lee et al., 2017), iPS cells are certainly supportive in
helping to better understand differentiation pathways of
stem cells and patient-specific bases of diseases, as well as
to develop personalized drug discovery efforts (c.f., e.g.,
Marsoner, Koch and Ladewig, 2018).
Dr. Paul Knoepfler and his team at UC Davis School of
Medicine (Davis, CA, USA) were the first to demonstrate
that induced pluripotency and oncogenic transformation
are related processes when comparing the transcriptomes
of iPS cells with the transcriptomes of cancer cells (Riggs
et al., 2013; see also Steinemann, Göhring and
Schlegelberger, 2013; Tung and Knoepfler, 2015).
Collectively, these studies have established a need for
caution when conducting clinical trials using iPS cell
based therapies.
2.3. Important differences regarding pluripotency
and multipotency between embryonic stem cells,
adult stem cells and cancer cells
There are important differences regarding totipotency,
pluripotency and multipotency between embryonic, fetal
and adult stem cells. A totipotent cell (found in the morula
stage of embryonic development) can differentiate into
any type of cell in the body. A pluripotent cell (found in
the blastocyst stage of embryonic development) possesses
the intrinsic pattern and existing guidelines for building a
new organ which is not yet existing. Pluripotency here
means that the embryonic stem cells can develop into any
tissue and organ of the three germ layers (i.e., ectoderm,
mesoderm and endoderm), without that these tissues or
organs yet exist.
In a similar way as in embryonic stem cells,
pluripotency is naturally present in adult stem cells as
evidenced by their ability to differentiate into ectodermal,
mesodermal and endodermal lineages. The difference
compared to embryonic stem cells, however, is that adult
stem cells cannot form new tissue on their own. They miss
the intrinsic program of embryonic stem cells but instead
depend on the signaling from the respective
microenvironment (outlined in detail in the following
sections). This means that adult stem cells may obtain any
of the three lineages but depend on constant induction of
differentiation and re-confirmation by signals released and
communicated from the local microenvironment (c.f., e.g.,
Rezza, Sennett and Rendl, 2014; Dong et al., 2015; Wabik
and Jones, 2015). If this information and confirmation is
missing or ceases, adult stem cells stop differentiating
(Trosko, 2014; Zhang et al., 2015).
In contrast, cancer cells with upregulated embryonic
genes like Oct4 can overcome this tissue
microenvironment dependency and form a tissue of their
initial tissue type, resulting in a metastasis (Tai et al.,
2004; Ilmer et al., 2014). For example, pancreatic tumor
cells can grow in the liver, forming a metastasis. In
contrast, normal pancreatic progenitor cells stop
differentiation and undergo apoptosis when implanted into
the liver. A continuous lineage specific differentiation
only occurs if it is supported by continuous signaling from
the respective microenvironment. For example, the
injection of cells derived from the bone marrow (which
for their majority consist of hematopoietic progenitor
cells) into a knee would result in a significant

5
inflammatory accumulation of white blood cells in the
knee. However, because the local microenvironment does
not support the continuing differentiation of these
hematopoietic progenitor cells, they would become
apoptotic. In fact, only true stem cells are able to continue
their expected differentiation pathway as supported by the
local microenvironment (e.g., Phinney and Prockop, 2007;
Kara et al., 2011; Malliaras et al., 2013). This is one of
several reasons why adult stem cell therapy with fresh,
unmodified autologous stem cells is very safe (e.g., Ra et
al., 2011; Hoffman and Dow, 2016; Toyserkani et al.,
2017). Safety concerns have only been raised with
cultured adult stem cells since as with higher passages
increased rates of potential malignant transformation may
occur (e.g. Røsland et al., 2009; Xu et al., 2012; Pan et al.,
2014).
3. The ubiquitous occurrence of stem
cells in tissue that contains blood
vessels
3.1. Location and origin of tissue resident stem
cells
Stem cells have been found and described at many sites in
the human body. However, many publications assume that
stem cells in specific organs are related and specific to the
respective tissue or organ. The hypothesis of universal,
vascular associated MSCs that are ubiquitously distributed
has been less adopted and accepted, both by the scientific
community and the public.
For many years MSCs were considered to be located
primarily in bone marrow. In recent years, however, stem
cells have been demonstrated to exist in many organs.
This is not so much due to the fact that each organ
contains its specific individual type of stem cells, but
rather that there is a ubiquitous distribution of the same
kind of MSCs throughout the human body.
Blood vessels are the initial structures to be formed if a
new organ is developing in an embryo. The presence of
stem cells in the vascular location allows equal
distribution of stem cells with pluripotent or multipotent
capacity throughout the body. These cells serve as a
repertoire for renewal of the respective tissue and organs
for the rest of the life of the individual.
For example, a certain number of MSCs per gram tissue
can be isolated from rat brain tissue. However, when
preparing a microvascular preparation of rat brain tissue –
in which only microvessels were remaining and the rest of
the brain tissue was discarded – we found that the
resulting number of MSCs per gram tissue increased by
several potencies, indicating that the majority of MSCs are
indeed located or associated with the vascular structure
(Alt et al.; unpublished results). Moreover, we were able
to demonstrate that the same vascular associated MSCs
can be isolated from blood vessels independent of the
organ or tissue they are derived from (outlined in detail in
Fig. 14 below).
3.2. Existence of mesenchymal stem cells in
blood vessels
Results of studies from our laboratories demonstrate that
identical vascular associated MSCs can be isolated from
all blood vessels, independent of the organ or tissue they
are derived from. This was demonstrated with cells
derived from both human and animal tissue. Specifically,
the MSCs were isolated from microvessels. Figure 3
shows a microvessel preparation and a phase contrast
image of such a microvessel from a rat brain.
FIGURE 3  Microvessel-associated cells (microvessel isolated
from a rat brain) represent a universal resource within the entire
body. (a) Scanning electron microscopic image of a microvessel.
(b) Phase contrast image of a microvessel as used for further
analysis. The scale bar represents 5 µm in (a) and 40 µm in (b).
Figure 4 shows a scanning electron microscopic image
of cells that were isolated from human abdominal adipose
tissue by enzymatic release, and were imaged just minutes
after isolation and plating. These cells (which represent
the so-called stromal vascular fraction) were not cultured
but freshly isolated. The presence of a composition of
different cells is found in this image. One can recognize
larger cells that exhibit a rough structure (‘P’ in Fig. 4).
These cells are progenitor cells that have already started to
build actin filaments (white arrows in Fig. 4) in order to
enhance their adherence to the extracellular matrix on
which they were seeded. Besides this, some of these cells
have started to communicate through microtubular
structures (arrowhead in Fig. 4). In addition there are also
lymphocytes (‘L’ in Fig. 4) as well as small cells (‘S’ in
Fig. 4) present. These small cells show a relatively smooth
surface and are limited in their ability to adhere to the

6
extracellular matrix. They do not show exosomal
structures on their surface as some of the larger cells do
(‘E’ in the high-power inset in Fig. 4). In addition, dying
cells are found which are devolving into apoptotic bodies
(‘D’ in Fig. 4).
FIGURE 4  Scanning electron microscopic image of the cellular
composition of the stromal vascular fraction obtained from
human abdominal adipose tissue. Abbreviations: P, progenitor
cells; S, small cells; D, dying cells; L, lymphocytes; E, exosomes.
The white arrows point to actin filaments, and the white
arrowhead to a micro-channel between two cells. Details are
provided in the main text. The scale bar represents 10 µm (20
µm in the high-power inset).
Analyzing the individual cells shown in Figure 4 in
more detail revealed that approximately 20% of the cells
were 'naked' and smooth on their surface, most likely
representing white blood cells such as lymphocytes (‘L’ in
Fig. 4). The content of the small assumed MSCs ('S' in
Fig. 4) was about 10% of the cell population shown in
Figure 4. This demonstrates that, following enzymatic
preparation, the stromal vascular fraction of human
abdominal adipose tissue is composed of white blood cells
and includes small MSCs and larger cells, most likely
progenitor cells. However, the composition between the
different cellular elements varies with the tissue these cells
are obtained from.
A more precise location of the cells in the vascular
structures is revealed by multi-colored
immunohistochemistry of a small arteriole (Fig. 5). Cell
nuclei (blue in Fig. 5) and smooth muscle antigen (SMA)
(green in Fig. 5) define the structure of the arteriole.
Furthermore, the location of laminin (purple in Fig. 5)
correlates with the border between the endothelial basal
lamina (representing the intima) and the media containing
the smooth muscle cells (note that in this position also the
internal elastic lamina is found; c.f., e.g., Di Russo et al.,
2017). Of note, neuron-glial antigen 2 (NG2) (red in Fig.
5) was found in close proximity to laminin, and most
likely represents the vascular associated MSCs.
FIGURE 5  Immunofluorescent detection of neuron-glial antigen 2 (NG2) (red in a,c,d; arrows), smooth muscle antigen (SMA) (green
in (b,c,d) and laminin (purple in d) in the wall of a small human arteriole (cell nuclei in blue). The scale bar represents 20 µm in (a,b) and
10 µm in (c,d).
NG2 has long been associated with pericytes (e.g.,
Crisan et al., 2012; Binamé, 2014; Sá da Bandeira,
Casamitjana and Crisan, 2017). However, pericytes are
fully differentiated cells that already have a terminal,
differentiated purpose in life, namely the formation of
capillaries together with endothelial cells (c.f., e.g., Avolio
and Madeddu, 2016). Thus, there are almost probably no
pericytes at the position of the NG2 signal in Figure 5, as

7
could be concluded from, e.g., Fig. 1B in Crisan et al.
(2012) (note that in Fig. 1 in Avolio and Madeddu (2016)
no pericytes were indicated at this position). Rather, it is
much more likely that the vascular associated MSCs are
also immunopositive for NG2 and, thus, NG2 is expressed
by more cells than just by pericytes. This would also
clearly contradict statements by, e.g., Crisan et al. (2012)
that pericytes and adventitial perivascular cells express in
situ and in culture markers of MSCs and display capacities
to differentiate towards osteogenic, adipogenic and
chondrogenic cell lineages, and by Sá da Bandeira,
Casamitjana and Crisan (2017) that pericytes are
precursors of MSCs that are capable of differentiating into
osteoblasts, adipocytes and chondrocytes.
In fact, different populations of pericytes and pericyte-
like progenitor cells were described. The best studied
pericyte is part of the typical capillary structure formed by
pericytes together with endothelial cells (e.g., Sims, 1986;
Diaz-Flores et al., 1991; Guillemin and Brew, 2004). The
second type is located at the surface of small blood
vessels, partly taking over vessel regulation (e.g.,
Fernández-Klett and Priller, 2015). Besides this, pericyte-
like progenitor cells were described in the adventitia of
larger vessels (c.f., e.g., Avolio and Madeddu, 2016).
However, all of these types of pericytes have positions in
the wall of capillaries or larger vessels that are clearly
distinct to the position of the NG2-positive cells shown in
Fig. 5. Other stainings and flow cytometric analyses
showed that the vascular associated MSCs are additionally
positive for Nestin (Fig. 6), Oct4 (Alt et al., 2011), Sca1
(Holmes and Stanford, 2007) and SSEA4 (D'Ippolito et
al., 2006).
FIGURE 6  Immunofluorescent detection of neuron-glial antigen 2 (NG2) (a), Nestin (b), CD29 (c), CD44 (d), CD146 (e), smooth
muscle antigen (SMA) (f), CD73 (g) and CD105 (h) in human cells that were freshly isolated from adipose tissue. Note the very small
cytosol of the cells that were immunopositive for NG2, Nestin and CD29 (a-c). The scale bar represents 20 µm in (a,e,f), 40 µm in (b),
26 µm in (c), 15 µm in (d), 18 µm in (g) and 36 µm in (h).
When cells that were freshly isolated from adipose
tissue are analyzed with immunohistochemistry there are
cells of different size, expressing different surface markers
(Fig. 6).
The very small cells express the aforementioned NG2
signal (Fig. 6a) as well as Nestin (Fig. 6b). Nestin is an
early marker of neural stem/progenitor cells as well as of
proliferative endothelial cells (Suzuki et al., 2010). These
cells have a tiny cytosol compared to the nucleus and to
other more differentiated cells. Also, cells immunopositive
for CD29 (integrin β1) exhibit this kind of small cytosolic
immunostaining (Fig. 6c). Together with CD49e (integrin
α-5) CD29 forms integrin α5β1, the primary receptor for
fibronectin (e.g., Clark, 1990; Serini, Valdembri and
Bussolino, 2006; Vega and Schwarzbauer, 2016). Most
probably the latter attaches the vascular associated MSCs
to the extracellular matrix within the vessel wall. This is
supported by the fact that SPARC (secreted protein acidic

8
and rich in cysteine; also known as osteonectin) (c.f., e.g.,
Lane and Sage, 1994; Murphy-Ullich and Sage, 2014) can
mobilize stem cells that are contained in adipose tissue
through its effect on integrin α5β1 (Nie et al., 2008; Tseng
and Kolonin, 2016), providing a functional basis for the
regulation of the contribution of these stem cells to tissue
and organ repair by SPARC. The latter is synthesized by
several types of cells, including osteoblasts and
odontoblasts (Holland et al., 1987), endothelial cells and
fibroblasts (Brekken and Sage, 2001), but also
macrophages (e.g., Chiodoni, Colombo and Sangaletti,
2010) and infiltrating leukocytes (Bradshaw, 2016). Thus,
SPARC may represent a key regulator in making the
vascular associated MSCs a replacement source
responsive to the signals of the surrounding tissue.
It is of note that SPARC is also expressed by ASCs in
vitro (Shaik et al., 2018). Moreover, the SPARC-related
modular calcium-binding protein 1 (SMOC1), a member
of the SPARC family and serving as a regulator of
osteoblast differentiation, was found in the secretome of
bone marrow derived MSCs (Choi et al., 2010). Thus,
SPARC may play a pivotal role in both affecting the
properties of MSCs in terms of proliferation and
differentiation based on cues from the extracellular
environment, as well as in paracrine activities of MSCs,
impacting upon the activities of other cells in the local
microenvironment (Kusuma et al., 2017).
3.3. The hypothesis of a universal, vascular
associated mesenchymal stem cell
Figures 7 and 8 illustrate the principle of location and
surface marker expression of the vascular associated
MSCs. Figure 7 shows the endothelial basal lamina in
black and the endothelial cells on the luminal side of the
endothelial basal lamina in gray. We postulate that the
vascular associated MSCs reside opposite to the
endothelial cells on the abluminal side of the endothelial
basal lamina, towards the smooth muscle cell layer and
embedded within those. This is in contrast to other
descriptions in the literature that vascular associated
MSCs, presumably immunopositive for pericyte markers,
would be located in the adventitia (e.g., Avolio and
Madudda, 2016; Corselli et al., 2012; de Souza et al.,
2016).
The vascular associated MSCs are small, which allows
them to migrate through tissue in order to help
maintaining tissue homeostasis. When these cells leave
their quiescent location (i.e., their primary niche depicted
in Fig. 8) they start to migrate, followed by proliferation
which is primarily under control of the Wnt signaling
pathway (e.g., Willert et al., 2003; Coudreuse and
Korswagen, 2007; Neth et al., 2007), and finally
differentiation. The location of proliferation may be called
the secondary niche (Fig. 8), in contrast to the primary
niche that is assigned to the vessel wall (Figs 7 and 8).
The cells can leave their secondary niche and differentiate
in the tertiary niche (Fig. 8) into their final lineage. The
tissue specific differentiation pathway is controlled by
signaling from the cells’ new microenvironment,
including micro-RNA and transcription factors (c.f., e.g.,
Rezza, Sennett and Rendl, 2014; Dong et al., 2015; Wabik
and Jones, 2015). This principle highlights the
regenerative power that exists in every organ. Throughout
life replacing cells exist that can be locally mobilized
upon need for tissue renewal and repair.
FIGURE 7  Schematic illustration of the hypothesized location
of the vascular associated MSCs (modified from Ilmer et al.,
2014).
3.4. Surface markers of mesenchymal stem cells
cultured in serum free media
Vascular associated MSCs can form spheroids when
cultured in special serum free media (c.f. Fig. 2). In this
regard it is crucial to bear in mind that only non-
differentiating MSCs are able to survive in serum free
media, while the majority of progenitor cells die.
Therefore, culturing cells in serum free media is a
powerful tool to discriminate between true MSCs,
progenitor cells and differentiated adult cells.
Figure 9 depicts the development of aggregation of
human ASCs in serum free media. After four days in
culture, individual cells aggregated and started to
communicate with each other, as shown in Figure 9a
where two cells are attached to each other (asterisk) and a
third cell forms a microtubular communicating structure
(arrow in Fig. 9a). A spheroid body with a diameter of
approximately 300 µm was formed after eight days in
culture (Fig. 9b). The shape of this spheroid body
resembled pretty much the shape of spheroid bodies
formed by embryonic stem cells, known as embryoid
bodies (e.g., Ader and Tanaka, 2014).

9
FIGURE 8  Expression of surface markers of mesenchymal stem cells in their primary, secondary and tertiary niche.
FIGURE 9  Aggregation of human adipose-derived stem cells
(a), leading to the formation of a spheroid (b) that resembles the
appearance of an embryoid body formed by embryonic stem
cells. The scale bar represents 20 µm in (a) and 75 µm in (b).
The understanding of the mode of differentiation of
MSCs and their surface markers has been confusing.
Specifically, markers such as CD44, CD73, CD90 and
CD105 – typically believed to be indicative of true stem
cells (Bourin et al., 2013) – are only present in cells that
have already left their silenced location and started to
enter the next developmental phase in order to attain
progenitor status. Besides this, CD44, CD90 and CD105
are also expressed in fibroblasts that exhibit no plasticity
at all (Alt et al., 2011).
We performed surface analysis of the antigen profiles of
human ASCs that were cultured for four days in serum
free media. Immunofluorescent analysis showed that these
cells were immunonegative for CD11b, CD14, CD31,
CD34, CD45 and HLA-DR (Fig. 10). Corresponding flow
cytometric analysis revealed that the relative numbers of
cells that were immunopositive for these surface markers
were smaller than 1.5% (relative number of cells that were
immunopositive for CD11b [CD11b+]: 1.1%; CD14+:
0.6%; CD31+: <0.1%; CD34+: 0.4%; CD45+: 1.3% and
HLA-DR: 0.2%). These results confirmed that cells that
express the following markers were not present in this cell
culture: macrophages (which typically are CD11b+),
hematopoietic progenitor cells (CD14+), endothelial
progenitor cells (CD31+), progenitor cells in general
(CD34+), cells expressing the pan-leukocyte marker
(CD45+) and cells expressing HLA-DR.

10
FIGURE 10  Immunofluorescent detection of CD11b (a), CD14 (b), CD31 (c), CD34 (d), CD45 (e) and HLA-DR (f) on the surface of
human adipose-derived stem cells that were cultured for four days in serum free media. The scale bar represents 60 µm in (a,d,f), 20
µm in (b), 40 µm in (c) and 30 µm in (f).
In contrast, Figure 11 shows the positive surface marker
antigen profile of human ASCs that were cultured for four
days in serum free media. Immunofluorescent analysis
showed that these cells were immunopositive for CD44,
CD73, CD90, CD105, Nestin and SMA. Corresponding
flow cytometric analysis revealed the following relative
numbers of immunopositive cells in this culture: relative
number of cells that were immunopositive for CD44
(CD44+): 92.4%, CD73+: 99.1%, CD90+: 28.8%,
CD105+: 73.3%, Nestin+: 76.2%, and SMA+: 41.2%.
Most importantly, in spheroids that were created from
unmodified human ASCs that were cultured for four days
in serum free media, cells expressed both Oct4 as an
indicator of 'stemness' as well as NG2. This clearly
indicates the natural presence of pluripotent cells in the
body without the need for prior genetical modification of
the cells as in case of iPS cells (Fig. 12).
For the sake of completeness it should be mentioned
that the results shown in Figs. 10-12 critically depend on
the time and mode of culturing. We demonstrated this by
culturing human ASCs in serum free media and analyzed
the cells after 24 hours and again after 120 hours. After 24
hours in culture, the relative numbers of CD11b+, CD34+,
CD45+ and CD105+ cells were 10.0%, 48.1%, 6.1% and
50.1%. In contrast, after 120 hours in culture the relative
numbers of CD11b+, CD34+ and CD45+ cells dropped to
less than 2%, whereas the relative number of CD105+
cells increased to 96.7%. We hypothesize that in culture
the loss of the CD34 surface marker as indicator of
progenitor status is a consequence of missing signaling
from the normally surrounding microenvironment in the
natural tissue.
3.5. The pluripotency of vascular associated
mesenchymal stem cell
It has been questioned whether a truly pluripotent stem
cell exists, or if the differentiation into the three germ
layers is based on the presence of a composition of
different progenitor cells that are responsible for the
individual differentiation capacity into the respective
lineage. To answer this question, we performed two key

11
experiments.
In the first experiment a single vascular associated
mesenchymal stem cell was clonally expanded for five
days in FBS, resulting in proliferation at a doubling time
of about 24 hours into millions of cells (Fig. 13a-d). Then,
cells were separated from this clonally expanded culture
and subjected to adipogenic, osteogenic, hepatogenic and
neurogenic induction media. It was found that the clonally
expanded cells, which all expressed the same genotypic
profile, were able to differentiate into ectoderm,
mesoderm and endoderm (Fig. 13e-h), indicating that the
initial single cell was indeed a true pluripotent stem cell.
FIGURE 11  Immunofluorescent detection of CD44 (a), CD73 (b), CD90 (c), CD105 (d), Nestin (e) and smooth muscle antigen (SMA)
(f) on the surface of human adipose-derived stem cells that were cultured for four days in serum free media. The scale bar represents
16 µm in (a,c), 12 µm in (b,d,f) and 20 µm in (e).
FIGURE 12  Immunofluorescent detection of Oct4 (a) and
neuron-glial antigen 2 (NG2) (b) in cells in spheroids that were
created by unmodified human adipose-derived stem cells that
were cultured for four days in serum free media. The scale bar
represents 20 µm.
In the second experiment we isolated vascular
associated MSCs from different organs (adipose tissue,
heart, skin, bone marrow and skeletal muscle) of rats and
subjected them after proliferation in FBS to adipogenic,
osteogenic, hepatogenic and neurogenic induction media.
Again, the cells were able to differentiate into ectoderm,
mesoderm and endoderm (Fig. 14).
The results of these key experiments provide significant
support for the hypothesis of a universal, vascular
associated mesenchymal stem cell with full pluripotency.

12
FIGURE 13  Clonal expansion of vascular associated human adipose-derived stem cells for five days in fetal bovine serum (FBS),
followed by multilineage differentiation induced by culturing the cells for three weeks in the respective induction media. (a-d) Culturing
of cells in FBS for 6 hours (a), 20 hours (b), 48 hours (c) and five days (d). (e-h) Induction of adipogenesis (mesoderm) (e),
osteogenesis (mesoderm) (f), hepatogenesis (endoderm) (g) and neurogenesis (ectoderm) (h), confirmed by Oil Red O staining (e) and
Alizarin Red S staining (f) as well as immunofluorescent detection of albumin (specific for hepatocytes) (g) and microtubule associated
protein 2 (MAP-2) (specific for neurons) (h). The scale bar represents 80 µm in (a-d), 100 µm in (e,f) and 40 µm in (g,h).
3.6. The three germ layer differentiation potential
of human adipose-derived stem cells
In 2007, we initially demonstrated the three germ layer
differentiation potential of human ASCs into adipocytes,
osteoblasts, hepatocytes and neurons (Bai et al., 2007).
While the cells cultured in non-inductive media did not
attain the lineage specific expression, cells subjected to
the specific induction media demonstrated the respective
differentiation (Fig. 15).
3.7. Regulation of the differentiation of stem cells
in a living organism
The first experiments we conducted in this regard were
carried out to highlight the possible influence of the
microenvironment surrounding the cells. This involved
co-culturing of neonatal rat cardiomyocytes with human
ASCs. In order to discriminate between the two different
types of cells, we labeled the human ASCs with green
fluorescent protein (GFP). Figure 16a shows a cell that is
immunopositive for the cardiac specific protein Troponin-
T (red signal). Figure 16b shows GFP (green signal) in the
cytosol of human ASCs. One cell shows a yellowish cell
body (arrow) that was obtained by the overlay of the green
signal (GFP) with a red signal originating from
immunofluorescent detection of MF20, one of the early
markers of the cardio-myogenic pathway differentiation
(c.f., e.g., Belema Bedada et al., 2005). In contrast, ASCs
that were not co-cultured with rat cardiomyocytes did not
show this early cardio-myogenic pathway differentiation.
Figure 17 shows a direct cell-cell contact between a rat
cardiomyocyte (identified by immunofluorescent detection
of Troponin-T; red signal) and a cell that expressed both
GFP and Troponin-T (yellowish signal). The latter
represents a human ASC at the latest stage of culturing. At
higher magnification (inset in Fig. 17) it appears that the
two cells adhered to each other, and reddish
microchannels appear to run from one cell to the other,
indicating a possible microchannel communication for the
exchange of information.
We further investigated the exchange of information
between ASCs and other cells by time-lapse video
microscopy of human ASCs that were labeled with red
Quantum dots (e.g., Yukawa et al., 2009) and were co-
cultured with MDA-MB-231 cells (a commercially
available human breast cancer cell line; c.f., e.g., Brinkley
et al., 1980) that were labeled with GFP (Fig. 18). In our
opinion the communication mechanism between cells

13
found in this experiment is the same as the cell-cell
connection of endothelial progenitor cells with cardiac
myocytes by nanotubes described by Koyanagi et al.
(2005) as an important mechanism for cell fate changes.
FIGURE 14  Induction of adipogenesis (mesoderm; confirmed by Oil Red O staining) (a,e,i,m,q), osteogenesis (mesoderm; confirmed
by Alizarin Red S staining) (b,f,j,n,r), hepatogenesis (endoderm; confirmed by immunofluorescent detection of albumin) (c,g,k,o,s) and
neurogenesis (ectoderm; confirmed by immunofluorescent detection of MAP-2) (d,h,l,p,t) by culturing rat vascular associated
mesenchymal stem cells obtained from adipose tissue (a-d), heart (e-h), skin (i-l), bone marrow (m-p) and skeletal muscle (q-t) for three
weeks in the respective induction media. The scale bar represents 100 µm in (a,e,m,q), 120 µm in (b), 50 µm in (c,d,f-h,j,n,r), 20 µm in
(i), 70 µm in (k), 30 µm in (l,p), 25 µm in (o) and 60 µm in (s,t).

14
FIGURE 15  Induction of adipogenesis (mesoderm; confirmed by Oil Red O staining) (a), osteogenesis (mesoderm; confirmed by
Alizarin Red S staining) (b), hepatogenesis (endoderm; confirmed by immunofluorescent detection of albumin) (c) and neurogenesis
(ectoderm; confirmed by immunofluorescent detection of MAP-2) (d) by culturing human adipose-derived stem cells (ASCs) for three
weeks in the respective induction media. (e-h) No induction of three germ layer differentiation by culturing human ASCs in non-inductive
control media. The scale bar represents 100 µm in (a,b,e,f) and 40 µm in (c,d,g,h).
FIGURE 16  (a) Immunofluorescent detection of Troponin-T in
the cytoplasm of a rat cardiomyocyte (red signal). (b) Human
adipose-derived stem cells that were labeled with green
fluorescent protein (GFP) (green signal). The arrow points to a
GFP-positive cell that was immunopositive for MF20 (resulting in
a yellowish signal). The scale bar represents 20 µm in (a) and 40
µm in (b).
FIGURE 17  Cell-cell contact between a rat cardiomyocyte
(expressing Troponin-T; red signal) and a cell expressing both
green fluorescent protein and Troponin-T (yellowish signal),
representing a human adipose-derived stem cell at the latest
stage of culturing (modified from Metzele et al., 2011). The inset
shows the contact between the cells at higher magnification. The
scale bar represents 15 µm (5 µm in the inset).

15
FIGURE 18  Time-lapse video microscopy of human adipose-derived stem cells (ASCs) that were co-cultured with MDA-MB-231 cells
labeled with green fluorescent protein (GFP) (green signal). One of the ASCs shown in this figure was labeled with a red Quantum dot
(red signal; the corresponding cell body is marked by a white arrow). In (f) this ASC formed a contact with another ASC (yellow arrows;
the cell-cell contact is indicated by the large yellow arrow in f). However, the Quantum dot was not exchanged between the cells. In (o)
the same ASC formed a contact with a MDA-MB-231 cell (green arrows; the cell-cell contact is indicated by the large green arrow in o)
and the Quantum dot was transferred from the ASC to the MDA-MB-231 cell. The high-power inset in o shows that the Quantum dot left
the ASC through a microchannel that was formed by the ASC itself, which demonstrates active participation of the ASC in exchange of
cellular components between the cells. The time interval between the frames was 40 min each. The scale bar represents 50 µm in a-t,
and 25 in the inset in o.
A second type of cell-cell communication is based on
genetic information contained in the exosomes (or
microsomes), which are released from the cell surface (c.f.
the inset in Fig. 4) and, upon uptake by pinocytosis (Fig.
19), induce a certain epigenetic reprogramming in the
recipient cell (e.g., Valadi et al., 2007; Xin et al., 2012;
Ismail et al., 2013). These exosomes are currently of
major interest with respect to further elucidating cell-cell
communication (e.g., Kruger et al., 2014). Specifically,
communication through the content of exosomes is
considered an important factor for orientation of cells (Xu
et al., 2017). Several micro-RNAs, transcription factors
and cytokines were identified to be involved in the
transfer of epigenetic reprogramming of cells in order to
direct them into a specific lineage differentiation. It was
also shown recently that microsomes obtained from a
murine pancreatic β-cell line are able to induce a beta cell
differentiation of bone marrow derived MSCs (Oh et al.,

16
2015), further supporting the notion that the orientation
and differentiation of MSCs are driven by signals of the
respective microenvironment (c.f. Valadi et al., 2007;
Ismail et al., 2013; Yamamoto et al., 2013).
Figure 19  Time-lapse video microscopy of human ASCs co-cultured with MDA-MB-231 cells labeled with GFP (green signal). The
ASCs were labeled with red Quantum dots (red signal). A Quantum dot in the cell body of one of the ASCs is marked by a red arrow in
(b-h); the corresponding cell is marked by a green arrow. The white arrows point to an exosome. The yellow arrow in (j) indicates the
event of pinocytosis of this exosome by the ASC that is marked by the green arrow. The time interval between the frames was 20 min
each. The scale bar in (t) represents 50 µm.
3.8. Immunosuppressive and anti-inflammatory
activities of stem cells
Both ASCs and bone marrow derived MSCs exhibit potent
immunosuppressive and anti-inflammatory activities (c.f.,
e.g., Krampera et al., 2003; González et al., 2009; Leto
Barone et al., 2013), and exosomes were shown to play an
important role in these processes (Caruso and Poon,
2018). In recent years apoptotic bodies, a major class of
extracellular vesicles released as a product of apoptotic
cell disassembly, have become recognized as another key
player in immune modulation (Caruso and Poon, 2018). A
recent study demonstrated that apoptosis in human bone
marrow derived MSCs induces recipient-mediated
immunomodulation in vivo (Galleu et al., 2017). On the
other hand, even in tissues with high cellular turnover,
apoptotic cells are rarely seen because of efficient
clearance mechanisms, including the sensing of cells that
undergo apoptosis via ‘find me’ signals (i.e., chemotactic
factors) (c.f., e.g., Hochreiter-Hufford and Ravichandran,
2013). One of these chemotactic factors is the

17
phospholipid, lysophosphatidylcholine (Lauber et al.,
2003). Increased concentration of lysophosphatidylcholine
was found in the medium in which hematopoietic
progenitor cells underwent apoptosis following growth
factor withdrawal (Fuchs et al., 2007). Our time-lapse
video microscopic investigations showed that human
ASCs that undergo apoptosis also stimulate the migration
of other cells to the apoptotic cell, and these other cells
can then take up the apoptotic bodies via pinocytosis (Fig.
20). However, it is currently unknown which ‘find me’
signals are used by ASCs that undergo apoptosis. In any
case, apoptosis of ASCs may be a key event in
immunosuppressive and anti-inflammatory activities
mediated by these cells.
FIGURE 20  Time-lapse video microscopy of human adipose-derived stem cells (ASCs) that were co-cultured with MDA-MB-231 cells
labeled with green fluorescent protein (GFP) (green signal; the red signal came from red Quantum dots that were exchanged between
the ASCs and the MDA-MB-231 cells (c.f. Fig. 18). The white arrow indicates an ASC that underwent apoptosis (named Cell A; the
disintegration of the cell body is marked in j by an asterisk). The red, yellow, green and blue arrows indicate four other cells (named
Cells B-E) that migrated towards Cell A during the three hours before disintegration of the cell body of Cell A (a-i). One of these cells
was an ASC (red arrows) whereas the other three cells were MDA-MB-231 cells. During the first 90 min after disintegration of the cell
body of Cell A, Cells B-E took up apoptotic bodies of Cell A by means of pinocytosis (indicated by larger arrows in the respective colors
in j,k,m,n). The time interval between the frames was 20 min each. The scale bar in (t) represents 50 µm.

18
3.9. A key role of stem cell therapy in re-
establishing tissue homeostasis between dying
and replacing cells
A key function of adult stem cells is to contribute to the
homeostasis of tissue resident parenchymal cells. As we
age, there is a continuous turnover in almost every tissue
between dying and replacing cells (with the exception of
some nerve cells in the brain, which will not be discussed
in detail here; c.f., e.g., Rakic, 1985; Breunig et al., 2007;
Jellinger and Attems, 2013). For a long time our body can
maintain tissue homeostasis; the equilibrium between
dying cells and stem cells is depicted in Figure 21a.
FIGURE 21  Homeostasis between dying cells (gray) and stem cells (orange) in tissue under healthy conditions (a), the situation
under pathological conditions and during aging (b), and after stem cell therapy (c). Under healthy conditions, tissue homeostasis is
maintained between dying cells and replacing stem cells. In contrast, tissue homeostasis is disrupted under pathological conditions and
during aging as more cells are dying than being replaced by stem cells. Application of a concentrated stem cell preparation can re-
establish tissue homeostasis.
However, tissue homeostasis can be disturbed with
increasing age in all tissues, such as tendons, bone, joints,
heart, liver, kidneys and muscles in a way that the
parenchymal cells, which are responsible for the organ
function, are more and more replaced by mesenchymal
fibroblastic cells. This is due to a lack of renewing power,
especially if ischemia, infections, accidents and other
inflammatory or traumatic events accelerate the tissue
turnover (Fig. 21b). A good example are chronic wounds
that show a number of problems, including insufficient
levels of cell proliferation, increased cell
senescence/apoptosis, impaired
angiogenesis/neovascularization, inflammation, increased
production of matrix metalloproteinases, increased matrix
degradation and decreased production of extracellular
matrix (c.f., e.g., Demidova-Rice, Hamblin and Herman,
2012).
An important question that has remained in this regard
is why injured organs (for example, a heart after a
myocardial infarction) do not source the stem cells from a
part of the body where they are present and would not be
‘missed’ after recruitment (such as from adipose tissue). It
is currently not possible to provide a satisfactory answer
to this question, because it would require to analyze model
organisms in which spontaneous recruitment of stem cells
from parts of the body where they are present and would
not be ‘missed’ after recruitment would occur. To our
knowledge, such model organisms are not known. Several
animal models for the study of limb regeneration have
been described, among them the Mexican axolotl
(Ambystoma mexicanum) (c.f., e.g., McCusker and
Gardiner, 2011; Farah et al., 2016; Grillo, Konstantinides
and Averof, 2016). It turned out that in limb regeneration,
a morphologically uniform intermediate (the so-called
blastema) is formed, consisting of a variety of multipotent
stem and progenitor cells originating from a variety of
tissues (Zielins et al., 2016). Further deciphering the
genetic and molecular regeneration inducers that are
involved in limb regeneration (Satoh, Mitogawa and
Makanae, 2015; Haas and Whited, 2017) may serve as the
basis to understand which signals could in general be used
by injured organs to recover stem cells from parts of the
body where they are not ‘missed’ after recruitment,
followed by investigations into why this does not happen
in the human body. In any case, the distance between the
stem cells in these parts of the body where they would not
be ‘missed’ after recruitment (as in adipose tissue) and
those cells that ‘call’ for the stem cells by the release of
cytokines may simply be too long.
Stem cell therapy is to be considered as the principal of
transferring concentrated stem cells, which have been
taken from one part of the body where they are not
‘missed’, to tissue in need of regeneration in order to re-
establish tissue homeostasis (Figure 21c). The isolation of
stem cells from suitable tissue (such as adipose tissue) and
their application to other injured tissue and organs can be
interpreted as the most gentle approach to help the body in
self-repair by increasing the numbers of stem cells at a

19
location where they are most needed. From these
considerations it also becomes clear that stem cell therapy
is not only directed to a specific organ, tissue or disease,
but will take the function of replacing and repairing tissue
and organs that suffer from a lack of repair, renewal and
rejuvenation.
4. Key advantages of adipose-derived
regenerative cells over bone marrow
derived stem cells for cell based
therapies
4.1. Comparison of bone marrow derived stem
cells with adipose-derived stem cells
For almost a decade bone marrow was the primary source
of MSCs for research into and development of therapies
based on adult stem cells. Bone marrow derived MSCs
exhibit significant potential for promoting tissue
regeneration, protection of ischemic tissue at risk, and
modulation of inflammation and autoimmunity (Murphy,
Moncivais and Caplan, 2013). However, utilizing bone
marrow derived MSCs for therapeutic purposes typically
requires to first isolate these cells and expand them in
culture. Because of the overwhelming presence of
hematopoietic progenitor cells in bone marrow that aim to
form new blood cells (Al-Drees et al., 2015; Ratajczak,
2015), only a small fraction of the cells in fresh bone
marrow aspirate are true pluripotent stem cells. In
contrast, other tissues such as adipose tissue yield orders
of magnitude more MSCs per unit volume than bone
marrow (e.g., Izadpanah et al., 2006; Yoshimura et al.,
2006; Bruno et al., 2014). In fact, as highlighted in the
previous sections of this article, adipose tissue is an organ
that is highly vascularized and contains a significant
number of MSCs. Thus, adipose tissue may be utilized as
a fresh cell preparation, rich in MSCs, without the need
for expansion in cell culture. There is also less damage to
the tissue where the stem cells are taken from, and
typically, the stem cells in adipose tissue have not been
challenged by ischemia, trauma or infections.
Compared to other sources of adult stem cells, adipose
tissue has the following specific advantages: (i) adipose
tissue is readily available in most individuals; (ii) small
amounts of adipose tissue (25 to 100 ml) can be harvested
using a simple liposuction procedure with low
invasiveness, with tolerable discomfort and low donor-site
damage; (iii) considerably larger amounts of MSCs can be
obtained from adipose tissue than from the same amount
of bone marrow; and (iv) the latter allows the usage of
ASCs without further need of culturing (named adipose-
derived regenerative cells [ADRCs] in order to
differentiate from cultured ASCs; c.f. Fig. 1). Given these
advantages, unmodified, autologous ADRCs (UA-
ADRCs) appear to be the most promising candidate for
repair and regeneration of many tissues, including chronic
wounds, soft tissue defects, bone and cartilage defects,
non-healing fractures, tendinopathies, diseased or injured
myocardium, urological conditions such as incontinence,
and neurological conditions (c.f., e.g., Schäffler and
Büchler, 2007; Gimble, Guilak and Bunnell, 2010; Altman
et al., 2010; Alt et al., 2011; Gir et al., 2012; Klein et al.,
2016).
4.2. Differences in the effectiveness of various
systems and methods that are available for
isolating adipose-derived regenerative cells
Different techniques and protocols were described for
releasing ADRCs for therapeutic use (c.f., e.g., Oberbauer
et al., 2015; Condé-Green et al., 2016; Van Dongen et al.,
2018). Collagenase I and II containing enzyme
preparations that degrade collagen are commonly used.
However, in order to release the vascular associated MSCs
from their binding site in the extracellular matrix inside
the blood vessels (and hereby to release the cells from
their 'hibernating' or silenced state) collagenases are only
partially effective. The addition of a neutral protease to a
collagenase enzyme preparation can significantly increase
the number of ADRCs recovered from a given volume of
adipose tissue. This was achieved by developing the
proprietary Matrase enzymatic reagent (InGeneron Inc.,
Houston, TX, USA). Isolating ADRCs with the Matrase
enzymatic reagent and the Transpose RT system
(InGeneron) appears advantagous to other commercial cell
separation systems (c.f. Haenel et al., 2018). Specifically,
ADRCs that were isolated with the Matrase enzymatic
reagent and the Transpose RT system may contain
approximately 40% of cells that are immunopositive for
CD29 and CD44, which are markers of ASCs (Haenel et
al., 2018). These authors also reported a colony forming
units (CFU) frequency (considered to be an indicator of
stemness) of approximately 11% of ADRCs isolated with
the Matrase enzymatic reagent and the Transpose RT
system. Metcalf et al. (2016) reported relative CFU values
of approximately 8% when isolating ADRCs from equine
adipose tissue. In contrast, relative CFU values between
0.2% and 1.7% were reported for ADRCs that were
isolated in head-to-head comparisons with four other
commercial cell separation systems (Aronowitz &
Ellenhorn, 2013; Aronowitz et al., 2016). However, a
direct comparison of the CFU values is hardly possible
due to significant methodological differences how the
respective numbers of CFUs were determined.

20
4.3. Long-term survival of adipose-derived stem
cells after transplantation in animal models into
the heart and subcutaneous locations
In order to study the capacity of ASCs to survive for a
long time at the site of engraftment, we transfected human
ASCs with a lentiviral vector expressing GFP and
luciferase (when the luciferase enzyme is expressed in
living cells, these cells are capable of converting
systemically injected luciferin dye into an active
luminescent fluorophor that can be detected
noninvasively; e.g., de Almeida, van Rappard and Wu,
2011; Scarfe et al., 2017). Figures 22 and 23 show the
time course of luciferin positive human ASCs that were
either intramyocardially delivered into severe combined
immunodeficient (SCID) mice after experimental
induction of myocardial infarction by permanent ligation
of the left anterior descending coronary artery (Fig. 22), or
into a subcutaneous location of SCID mice (Fig. 23),
respectively. Strong bioluminescence signals were found
over the injection sites at all investigated time points
(indicated in Figs 22 and 23), demonstrating the long-term
survival of ASCs delivered into injured hearts or at
subcutaneous locations, respectively. At no other location
of the body of the SCID mice a cell engraftment was
detected.
To better understand the fate of the delivered ASCs, we
investigated subcutaneous tissue harvested at the injection
site at four weeks after subcutaneous application of human
ASCs. Immunofluorescent detection of von Willebrand
factor (red signal in Fig. 23h) and Lamin A/C (green
signal in Fig. 23h) demonstrated that the human ASCs
delivered into a subcutaneous location of a SCID mouse
participated in the formation of new blood vessels.
FIGURE 22  In vivo bioluminescence imaging over time of 5×105 human adipose-derived stem cells (ASCs) that were
intramyocardially delivered into SCID mice after experimental induction of myocardial infarction by permanent ligation of the left anterior
descending coronary artery. The ASCs were transfected with a lentiviral vector expressing green fluorescent protein and luciferase.
Strong bioluminescence signals were found over the heart region at all investigated time points (arrows), indicating the long-term
survival of delivered ASCs in injured hearts (modified from Bai et al., 2011).
4.4. Specific therapeutic benefits of adipose-
derived regenerative cells
One of the most striking features of ADRCs in cell-based
therapy is their pluripotent differentiation potential
without any prior manipulation, genetic alteration or the
need for culturing. Also, the use of ADRCs in cell-based
therapies has a number of additional benefits. Specifically,
ADRCs can be (i) obtained from a small amount of
adipose tissue when using the appropriate technology for
isolating the cells; and (ii) harvested from and re-applied
to the same patient at the point of care without the need
for expensive equipment, complicated processing or

21
repeat interventions.
It is crucial to bear in mind that in contrast to ASCs,
fresh, uncultured ADRCs in principle cannot be labeled.
Accordingly, it is not possible to experimentally determine
whether the following benefits of ASCs also apply to
ADRCs (although it is reasonable to hypothesize that this
is indeed the case). Specifically, ASCs can (i) stay locally,
survive and engraft in the new host tissue into which the
cells were applied; (ii) differentiate under guidance of the
new microenvironment into cells of all three germ layers
(c.f. Figs 13-15); (iii) integrate into and communicate
within the new host tissue by forming direct cell-cell
contacts (c.f. Fig. 17); (iv) exchange genetic and
epigenetic information through release of exosomes (Fig.
18); (v) participate in building new vascular structures in
the host tissue (c.f. Fig. 23h and, e.g., Haenel et al., 2018);
(vi) positively influence the new host tissue by release of
cytokines (among them vascular endothelial growth factor
and insulin-like growth factor 1) (Sadat et al., 2007); (vii)
protect cells at risk in the new host tissue from undergoing
apoptosis (Sadat et al., 2007); and (viii) induce immune-
modulatory and anti-inflammatory properties (c.f., e.g.,
González et al., 2009; Leto Barone et al., 2013), whereby
the inhibiting effect on apoptosis may play an important
role (Fig. 20).
FIGURE 23  (a-g) In vivo bioluminescence imaging over time of human adipose-derived stem cells (ASCs) that were delivered
together with crosslinked collagen into a subcutaneous location of SCID mice. The ASCs were transfected with a lentiviral vector
expressing green fluorescent protein and luciferase. Bioluminescence signals were found at the injection site up to 6 month (arrows),
indicating the long-term survival of delivered ASCs in subcutaneous tissue (h). Immunofluorescent detection of von Willebrand factor
(vWF; red signal) and Lamin A/C (green signal) as indicator of human cells in a SCID mouse, forming a blood vessel four weeks after
application of the human ASCs into a subcutaneous wound (counterstaining with DAPI; blue signal). The scale bar represents 20 µm in
h.
5. Examples of application of unmodified,
autologous adipose-derived regenerative
cells (UA-ADRCs) in the praxis of
regenerative cell therapy
5.1. Should UA-ADRCs be applied locally or
systemically?
It has been shown from experimental data that local
injection of stem cells is safe (Altman et al., 2011) as
typically a local inflammatory trigger retains stem cells at
the site of injection. Meta analyses of clinical results as
well have confirmed the safety of stem cell therapy in
various indications (Lalu et al. 2012, Ra et al., 2011,
Hoffmann and Dow, 2016).
When stem cells are injected into the circulation, they
are ‘searching’ for a place where they could be of benefit.
As a tumor is considered a 'wound that does not heal'
(Flier, Underhill and Dvorak, 1986; Dvorak, 2015), it
releases cytokines and other factors that aim to attract
circulating stem cells to the tumor site (c.f., e.g., Gehmert
et al., 2010; Ilmer et al., 2014; Kruger et al., 2014), where
stem cells may assist the tumor to build its stroma and
thereby help the tumor to grow faster (Altman et al.,
2009). Hence, UA-ADRCs preferably should be applied
locally to the side of need. In case of systemic application
the potential of UA-ADRCs to support an already existing
tumor in its growth (in contrast to the absent ability of
UA-ADRCs to induce a de novo tumor) should be
considered, pointing to the need for evaluation of the
oncogenic status of the patient prior to a systemic
application.
5.2. Cartilage defects
The advantages of treating cartilage defects with UA-
ADRCs are exemplified by the following example of a
male, 51-year-old patient who presented with recurring
and increasing pain in both knee joints during walking and

22
other activities (all treatments and procedures described in
this section were performed in the framework of a clinical
study that was approved by the Freiburg Ethics
Commission International (feki; Freiburg, Germany) (feki
code 013/1371)). The patient's history included a tibial
chondrocyte transplant that had been performed three
years previously. Figure 24a shows an arthroscopic view
of third-degree damage to the right tibial plateau where
the transplanted chondrocytes were gone and only the
artificial matrix with small holes implanted on the tibial
plateau was still present (white asterisk in Fig. 24a).
Furthermore, considerable osteoarthritic damage of the
femoral cartilage was observed (black asterisk in Fig.
24a). Figure 24b shows the situation after arthroscopic
removal of the failed chondrocyte transplant (white
asterisk in Fig. 24b) as well as 'mushy' and damaged
cartilage structure on the femoral condyles before it was
removed (black asterisk in Fig. 24b). Then, the right knee
was treated with a single application of UA-ADRCs
obtained from 50 g of abdominal adipose tissue. A control
arthroscopy one year later showed complete healing of the
tibial defect (white asterisk in Fig. 24c) and of the femoral
parts, with formation of new whitish cartilage that showed
a sharp demarcation border to the original, more yellowish
cartilage (arrows in Fig. 24c). The left knee of the same
patient was treated with a standard procedure without
application of UA-ADRCs, i.e., arthroscopic removal of
damaged cartilage and drilling of small holes into the
bone. A control arthroscopy one year later showed a
somewhat uneven, overshooting fibroblastic scar
formation (asterisk in Fig. 24d) without a sharp
demarcation border to the original cartilage (arrows in Fig.
24d). This indicated that there was some sort of healing,
but not a regrowth of organized cartilage, as we
hypothesize for the right knee after application of UA-
ADRCs.
Figure 25 shows arthroscopic views of cartilage defects
of the knees of two other patients that were also
successfully treated with UA-ADRCs. Of note, the finding
of a sharp demarcation border between the newly formed
cartilage and the original cartilage (arrows in Fig. 24c)
was found as well in these patients one year after
treatment with UA-ADRCs (arrows in Fig. 25b,d). Again,
we hypothesized that this arthroscopic finding indicated
regrowth of organized cartilage.
To test the hypothesis that the sharp demarcation
borders between the newly and the original cartilage
(arrows in Fig. 24c and Fig. 25b,d) indicated regrowth of
organized cartilage after application of UA-ADRCs, we
obtained written, informed consent by the patient
represented in Fig. 24 that small tissue samples could be
taken from the regenerated tissue during the follow-up
arthroscopy. Histologic analysis of the tissue samples
showed two key differences between the samples: (i) After
application of UA-ADRCs the newly formed cartilage
showed (like in a textbook of histology; c.f., e.g., Jung,
2014) a zonal organization with differently shaped
chondrocytes in a superficial, middle and deep layer (Fig.
26). In contrast, without application of UA-ADRCs a
more amorphous fibrocartilage with scattered cells
(arrows in Fig. 26b) was achieved that had no such
layered organization (Fig. 26b).
FIGURE 24  Example of successful application of unmodified
autologous adipose-derived regenerative cells (UA-ADRCs) for
treating cartilage defects. The panels show arthroscopic views of
the right (a-c) and the left (d) knee of a male, 51-year-old patient
who presented with recurring and increasing pain in both knee
joints during walking and other activities. (a) Third-degree
damage to the right tibial plateau (white asterisk) after a tibial
chondrocyte transplantation that had been performed three years
previously, as well as considerable osteoarthritic damage of the
femoral cartilage (black asterisk). (b) Situation of the right knee
after arthroscopic removal of the failed chondrocyte transplant
(white asterisk) as well as 'mushy' and damaged cartilage
structure on the femoral condyles before it was removed (black
asterisk). (c) Situation of the right knee one year after
arthroscopic removal of damaged cartilage and a single
application of UA-ADRCs isolated from 100 ml lipoaspirate,
showing complete healing of the tibial defect (white asterisk) and
of the femoral parts, with formation of new whitish cartilage that
shows a sharp demarcation border to the existing old and more
yellowish cartilage (arrows). (d) Situation of the left knee one
year after performing a standard procedure without application of
UA-ADRCs (i.e., arthroscopic removal of damaged cartilage and
drilling of small holes into the bone), showing a somewhat
uneven, overshooting fibroblastic scar formation (black asterisk)
without a sharp demarcation border to the original cartilage
(arrows). Abbreviations: R, right; L, left; BL, baseline; M12,
twelve months after baseline.

23
FIGURE 25  Other examples of successful application of
unmodified autologous adipose-derived regenerative cells (UA-
ADRCs) for treating cartilage defects. The panels show
arthroscopic views of the knees of two patients (Patient 1 [male,
45 years old]: a,b; Patient 2 [male, 55 years old]: c,d) before (a,
c) and one year after (b,d) arthroscopic removal of damaged
cartilage (asterisks in a,c) and application of UA-ADRCs isolated
from 50 ml lipoaspirate. The arrows in (b,d) indicate the sharp
demarcation border between the newly formed and the original
cartilage. The white arrow in (a) points to the arthroscopic
instrument that was used to remove damaged cartilage.
(ii) Furthermore, after application of UA-ADRCs the
contact zone between the newly formed cartilage and bone
showed (also like in a textbook of histology) typical
chondrocytes with a small nucleus and a hollow space
around (arrows in Fig. 26c). In contrast, without
application of UA-ADRCs the contact zone between the
newly formed cartilage and bone showed an infiltration
with inflammatory cells, fibroblasts (arrows in Fig. 26d)
and small blood vessels (arrowheads in Fig. 26d).
Analysis of the tissue sample taken during arthroscopic
inspection of the right knee of the patient represented in
Fig. 24 at one year after arthroscopic removal of damaged
cartilage and a single application of UA-ADRCs from 50
g of his adipose tissue with polarized light microscopy
demonstrated that the collagen fiber bundles in the deep
and middle layers of the newly formed cartilage show a
more vertical orientation (i.e., perpendicular to the border
between bone and cartilage), whereas the collagen fiber
bundles in the superficial layer showed a more horizontal
orientation (i.e., parallel to the surface) (Fig. 27). This
finding is in line with the description of the physiologic
orientation of collagen fiber bundles in articular cartilage
in the literature when analyzed with polarized light
microscopy (e.g., Rieppo et al., 2008).
To our knowledge, the results presented in this section
go beyond the state-of-the-art in the field of regenerating
damaged cartilage with ADRCs and ASCs. In a recent
review (Damia et al., 2018) a number of clinical studies
were listed in which cartilage defects in the human knee
were treated with ASCs (Jo et al., 2014; Pers et al., 2016;
Freitag et al., 2017). Of note, in all of these studies ASCs
were applied, whereas we have been using fresh
uncultured ADRCs (for the advantages of ADRCs over
ASCs see Sections 4.1. and 4.4.). The maximum follow-
up period in the studies by Jo et al. (2014) and Pers et al.
(2016) was only six months after application of ASCs
(MRI, arthroscopy and histologic analysis in both studies;
n=18 patients in both studies). In the single case report by
Freitag et al. (2017) MRI was performed at twelve months
after application of ASCs, but no arthroscopy and, thus, no
histologic analysis. Furthermore, in none of these studies
tissue samples were ever investigated with polarized light
microscopy.
5.3. Chronic, recalcitrant low back pain caused
by lumbosacral facet syndrome
Lumbosacral facet syndrome is a term used to describe a
painful condition caused by inflammation and irritation of
the zygapophyseal (facet) joints of the spine (Fig. 28a-c).
It is most commonly caused by chronic degenerative
changes in the lumbosacral spine. Symptoms include low
back pain with or without referral to the lower extremities
(Alexander and Dulebohn, 2018). To our knowledge,
reports on cell-based therapies for chronic low back pain
caused by lumbosacral facet syndrome have not yet been
published.
Figure 28d-f shows typical wear and tear that occurred
in the spine of three former professional ski racers (all
treatments and procedures described in this section were
performed in the framework of clinical assessment). The
green arrows indicate normal intervertebral disc structure
between the lumbar vertebrae L3 and L4, demonstrating
that there was a certain level of water content, reflected by
the white signal in the MRI. In contrast, the yellow arrows
show that between vertebrae L4 and L5 in all three
athletes there was a reduction in height of the
intervertebral disc due to the diminished overall volume of
the intervertebral disc. The increased body weight now
resting on the facet joints results in an inflammatory
reaction of the facet joints. All three athletes were treated
with UA-ADRCs. The cells were injected into the right
and left facet joints between L4 and L5 as well as between
L5 and S1 within a single procedure of harvesting and
injection that lasted about two hours. This treatment
resulted in a significant and long lasting (now more than
three years) pain reduction that enabled these athletes to
return to successful competitive sports.

24
FIGURE 26 Representative photomicrographs of 5 µm thick histologic sections (stained with toluidin blue [a,b] or hematoxylin and
eosin stain [c,d]) of the small tissue samples taken during arthroscopic inspection of the knees of the patient represented in Fig. 24 at
one year after arthroscopic removal of damaged cartilage and a single application of unmodified autologous adipose-derived
regenerative cells (UA-ADRCs) (right knee) (a,c) or after performing a standard procedure without application of UA-ADRCs (i.e.,
arthroscopic removal of damaged cartilage and drilling of small holes into the bone) (left knee) (b,d), respectively. The dotted lines in (a)
indicate a zonal organization of the newly formed cartilage with differently shaped chondrocytes in a superficial (SL), middle (ML) and
deep layer (DL). The arrows in (b) point to scattered cells within newly formed amorphous fibrocartilage. The arrows in (c) indicate
typical chondrocytes with a small nucleus and a hollow space around in the contact zone between the newly formed cartilage and bone,
whereas the arrows in (d) point to an infiltration with inflammatory cells in the contact zone between the newly formed cartilage and
bone. The arrowheads in (d) indicate small blood vessels. The scale bar represents 100 µm.
Figure 29 shows individual pain scores on a VAS scale
(with 0 representing no pain and 10 representing
maximum, unbearable pain) of n=39 patients with chronic,
recalcitrant low back pain caused by lumbosacral facet
syndrome before (red dots in Fig. 29) and one year after
(green dots in Fig. 29) treatment with UA-ADRCs isolated
from 100 ml lipoaspirate each (the follow-up interval
ranged between twelve months and more than three years)
(modified from Rothoerl et al., 2016) (all treatments and
procedures described in this section were performed in the
framework of a clinical study that was approved by the
Freiburg Ethics Commission International (feki; Freiburg,
Germany) (feki code 016/1252)). The mean and standard
error of the mean of the VAS scores before and after
treatment was 7.21 ± 0.17 and 1.80 ± 0.17; this difference
was highly statistically significant (Wilcoxon matched-
pairs signed rank test; p < 0.001). As a consequence, the
quality of life of these patients was significantly
improved.
Figure 30 shows mean and standard deviation of the
individual improvement of the VAS score after treatment
as a function of the VAS score before treatment of these
39 patients. Linear regression analysis showed a
statistically significant relationship between the VAS
score before treatment and the individual improvement of
the VAS score after treatment, with the best results
obtained with the highest VAS scores before treatment (r2
= -0.22; p = 0.003). This impressively demonstrates the
potential of UA-ADRCs for the treatment of chronic,
recalcitrant low back pain caused by lumbosacral facet
syndrome, and perhaps opens the possibility for obtaining
comparable results for other chronic pain conditions of the
musculoskeletal system.

25
FIGURE 27  Representative photomicrographs of 5 µm thick
histologic sections (stained with hematoxylin and eosin stain) of
the small tissue sample taken during arthroscopic inspection of
the right knee of the patient represented in Fig. 24 at one year
after arthroscopic removal of damaged cartilage and a single
application of the unmodified autologous adipose-derived
regenerative cells. The panels (b,d) were generated with
polarized light microscopy and converted to grayscale. The red
lines in (b,d) indicate the orientation of the collagen fiber bundles
within the newly formed cartilage (more vertically in in the deep
and middle layers, and more horizontally in the superficial layer).
The asterisk in (c) indicates a region of original cartilage,
demonstrating that the tissue sample was indeed taken from the
newly formed cartilage. The scale bar represents 100 µm.
5.4. Guided bone regeneration
The promising results achieved in the treatment of
cartilage defects and chronic, recalcitrant low back pain
caused by lumbosacral facet syndrome (described in detail
in Sections 5.1. and 5.2.) gave reason to hypothesize that
the application of UA-ADRCs could also advance the
treatment of other pathologies of the musculoskeletal
system. In this regard it was a key finding that in guided
bone regeneration (GBR) (exemplified by a case of a 79-
year old patient who presented with a partly failing
maxillary dentition and who was treated with a bilateral
external sinus lift procedure as well as a bilateral lateral
alveolar ridge augmentation) the combined application of
UA-ADRCs, Fraction 2 of plasma rich in growth factors
(PRGF-2) and an osteoinductive scaffold (OIS)
(Treatment A) was superior to the combination of PRGF-2
and the same OIS alone (Treatment B) (Solakoglu et al.,
2019) (these treatments and procedures were performed in
the framework of a clinical study that was approved by the
Ethics Committee of the Federal Dental Association
Hamburg (Hamburg, Germany) (no. PV5211)).
Specifically, Treatment A resulted in faster build up of
higher relative amounts (area/area) of newly formed bone,
connective tissue and arteries as well as in lower relative
amounts of adipocytes and veins at 34 weeks after GBR
(Fig. 31a,b) comparted to standard treatment without stem
cells.
Avascular necrosis of the femoral head is one of the
many indications where bone regeneration is essential for
rehabilitation (e.g., Larson et al., 2018). Cell-based
therapy for this pathology has been addressed in a number
of clinical studies (for a recent systematic review and
meta-analysis see, e.g., Andriolo et al., 2018). One of
these studies was a case report of a 43-year-old male
patient who was successfully treated with ADRCs mixed
with platelet-rich plasma and hyaluronic acid (Pak et al.,
2014). We went one step further and treated a 41-year-old
male patient suffering from avascular necrosis of the
femoral head only with ADRCs (all treatments and
procedures were performed in the framework of clinical
assessment). Figure 31c shows an MRI scan of the left hip
of this patient who was confined to a wheelchair because
of unbearable pain during walking, with a significant
necrotic space in the head of the patient’s left femur. Six
months after two applications of UA-ADRCs (that were
locally injected through a channel which came through the
lateral side of the greater trochanter) a control MRI scan
showed a markedly improved situation (Fig. 31d), and the
patient could leave the wheelchair and walks now without
pain.

26
FIGURE 28  (a-c) Schematic of the lumbar vertebrae L3-L5 and the upper edge of the os sacrum of a human spine from dorsal (a)
and lateral (b,c). The red arrows indicate the zygapophyseal (facet) joints between L3 and L4 (L3/L4), between L4 and L5 (L4/L5) and
between L5 and the os sacrum (L5/S1) with joint capsule (on the left side in a and in b) as well as without joint capsule (on the right side
in a and in c). The green arrows in (b,c) indicate the intervertebral disc between L3 and L4, and the yellow arrows the intervertebral disc
between L4 and L5. (d-f) MRI scans of the lumbar spine (from lateral) of three former professional, internationally highly successful ski
racers (we are not allowed to provide more information about these patients in order to protect privacy). The green arrows indicate
normal intervertebral disc structure between L3 and L4, whereas the yellow arrows show that in all three athletes there was a reduction
in the height of the intervertebral disc (and, thus, the intervertebral space) between L4 and L5. The asterisk in (f) points to a bone
marrow edema in the vertebral body of L5, and the black arrow in (f) to an upper plate collapse of L5. All three athletes were treated
with unmodified autologous adipose-derived regenerative cells. The cells were injected into the right and left facet joints between L4
and L5 as well as between L5 and S1 within a single procedure of harvesting and injection that lasted about two hours. This treatment
resulted in a significant and long lasting (now more than three years) pain reduction that enabled these athletes to return to successful
competitive sports.

27
FIGURE 29  Individual VAS scores of n=39 patients with chronic, recalcitrant low back pain caused by lumbosacral facet syndrome
before (red dots) and after (green dots) treatment with UA-ADRCs isolated from 100 ml lipoaspirate each (the follow-up data ranged
between six months and three years) (modified from Rothoerl et al., 2016). The patients were listed consecutively according to the date
of treatment.
FIGURE 30  (a,b) Representative photomicrographs of 3 µm
thick paraffin sections stained with hematoxylin and eosin stain
of biopsies that were collected at 34 weeks after performing
guided bone regeneration (in the framework of a bilateral
external sinus lift procedure as well as a bilateral lateral alveolar
ridge augmentation in a 79-year old patient who presented with a
partly failing maxillary dentition) using a combination of
approximately 50×106 unmodified autologous adipose-derived
regenerative cells (UA-ADRCs), Fraction 2 of plasma rich in
growth factors (PRGF-2) and an osteoinductive scaffold (OIS)
(right side) (a) or a combination of PRGF-2 and the same OIS
alone (left side) (b). Abbreviations: A, adipocytes; B, newly
formed bone; S, scaffold; V, veins. The scale bar represents 100
µm. (c,d) MRI scans of the left hip of a 41-year-old male patient
suffering from avascular necrosis of the femoral head (arrows in
c,d) at baseline (c) and at six months after two applications of
UA-ADRCs.
FIGURE 31  (a,b) Representative photomicrographs of 3 µm
thick paraffin sections stained with hematoxylin and eosin stain
of biopsies that were collected at 34 weeks after performing
guided bone regeneration (in the framework of a bilateral
external sinus lift procedure as well as a bilateral lateral alveolar
ridge augmentation in a 79-year old patient who presented with a
partly failing maxillary dentition) using a combination of
approximately 50×106 unmodified autologous adipose-derived
regenerative cells (UA-ADRCs), Fraction 2 of plasma rich in
growth factors (PRGF-2) and an osteoinductive scaffold (OIS)
(right side) (a) or a combination of PRGF-2 and the same OIS
alone (left side) (b). Abbreviations: A, adipocytes; B, newly
formed bone; S, scaffold; V, veins. The scale bar represents 100
µm. (c,d) MRI scans of the left hip of a 41-year-old male patient
suffering from avascular necrosis of the femoral head (arrows in
c,d) at baseline (c) and at six months after two applications of
UA-ADRCs.

28
5.5. Examples of other treatments with UA-
ADRCs
Several studies on animal models and clinical pilot studies
have shown that human ADRCs and ASCs are able to
enhance and accelerate wound healing, especially in
chronic wounds (e.g., Altman et al., 2009; Sorice et al.,
2016; Konstantinow et al., 2017). An example of
successful application of UA-ADRCs for treating chronic
wounds in humans is shown in Figure 32, which represent
cases of prolonged wounds that did not heal for several
years (these treatments and procedures were performed in
the framework of a clinical study that was approved by the
Ethics Committee of the Medical Faculty of the Technical
University Munich (Munich, Germany) (no. 5639/12)).
The underlying principle in those wounds is that the
regenerative power of the local stem cells is exhausted due
to the repeated futile attempts at healing. Debridement of
the wounds showed that debrided tissue contained only
very few living cells. The transfer of UA-ADRCs
activated the wounds after a few days and induced
shrinkage and significant reduction of the inflammatory
redness. Typically, after two to three months all wounds
based on venous insufficiency were closed. Of note, the
images shown in Figure 32 impressively demonstrate that
UA-ADRCs do not only work in younger people. In one
85-year-old patient who suffered from open leg wounds of
more than 100 cm2 for several years, healing was obtained
after a single application of UA-ADRCs (Fig. 32c).
Primarily not considered a mainstream indication for
stem cell therapy, there is anecdotal evidence indicating
that UA-ADRCs have a great effect on remodeling of scar
tissue. As demonstrated in Figure 33, the degradation of
the extracellular matrix, which leads to scarring, can be
reversed (all treatments and procedures described in this
section were performed in the framework of clinical
assessment). We still do not fully understand the exact
mechanisms behind this effect (Altman et al., 2009; 2010)
However, the expression of matrix metalloproteinases 2
and 9, which enable MSCs to migrate through tissue (e.g.,
Pourjafar et al., 2017), might be responsible for the
realignment of the extracellular matrix. We hypothesize
that scar tissue becomes primarily decomposed and
degraded by MSCs when they migrate through tissue, and
afterwards, the extracellular matrix is recomposed and
restructured by a part of the posterior cell body of the
MSCs (c.f., e.g., Te Boekhorst, Preziosi and Friedl, 2016).
Figure 34 shows the effects of UA-ADRCs on hair
growth (all treatments and procedures described in this
section were performed in the framework of clinical
assessment). The mechanism is assumed to be similar to
the regeneration of an organ or other tissues. Specifically,
we hypothesize that the delivered stem cells most likely
'are guided' by the remaining hair roots to their 'local task'
of trans-differentiation into hair.
6. Outlook
Regenerative medicine and cell therapy are not yet part of
mainstream clinical practice. Therapies based on UA-
ADRCs discussed in this article appear to be highly
promising candidates for repair and regeneration of many
tissues and ultimately for wide adoption to the practice of
medicine. One of the most striking features of UA-
ADRCs is their pluripotent differentiation potential
without any prior modification or need for culturing.
Furthermore, UA-ADRCs can be obtained from a small
amount of adipose tissue when using the appropriate,
enzyme-supported technology for isolating true
pluripotent stem cells. The fact that tissue can be
harvested from and cells can be re-applied to the same
patient at the point of care in one clinical session without
the need for expensive equipment, complicated processing
or repeat interventions indicates easy integration into the
clinical workflow.
As with any medical innovation, the scientific and
medical community interested in these novel therapies
needs to develop sound clinical evidence to further show
safety and efficacy of cell-based therapies. Our
understanding of the mechanism of actions and potential
benefit of stem cell therapy has increased enormously over
the past decade and we hope that there is now enough data
to convince others to embark on scientifically-designed
clinical studies that will provide the necessary objective
evidence. Especially musculoskeletal indications with
their large incidence and prevalence rates and often
substantive total cost of care associated with current
clinical practice should prove to be attractive candidates
for such efforts.
An important factor for successful implementation of
autologous cell therapies will be the proactive support of
regulatory authorities to design frameworks that, while
addressing valid concerns around the safety of unproven
therapies that can be found in some places currently, show
a clear path to market approval and reimbursement. Some
progress in that direction can be observed in a number of
jurisdictions. For example, the US FDA has provided clear
guidance for Regenerative Medicine Advanced Therapies
in 2017. Other countries such as Japan and Australia also
seem to have created an advanced regulatory framework
that will allow patients to gain fast access to Regenerative
Medicine. Unfortunately, the situation in the European
Union (EU) is not yet that clear. A complex set of
guidance and regulations that result in somewhat
convoluted responsibilities between the EU, Member
States and local authorities will need to eventually be
streamlined to help to make these promising therapies

29
available for all patients in the EU too.
FIGURE 32  Examples of successful application of unmodified autologous adipose-derived regenerative cells (UA-ADRCs) for treating
chronic wounds that did not heal for years. (a-d) Female, 76 year old patient; chronic wound over the medial malleolus; single
application of UA-ADRCs isolated from 30 ml lipoaspirate. (e-h) Male, 82 years old patient; chronic wound on the lower leg caused by a
phosphorus bomb during World War II; single application of UA-ADRCs isolated from 30 ml lipoaspirate. (i-l) Female, 84 years old
patient, chronic wound on the lower leg; single application of UA-ADRCs isolated from 30 ml lipoaspirate. Abbreviations: BL, baseline;
D9, nine days after application of UA-ADRCs; M1/M2/M2.5/M3/M5/M7, one/two/two and a half/ three/five/seven months after
application of UA-ADRCs.
It is our hope that this article will help to clarify and
facilitate the understanding, development and adoption of
Regenerative Medicine in general and specifically of
therapies based on UA-ADRCs. Not all medical
challenges that we face in clinical practice today will be
candidates for Regenerative Medicine. But as we are
facing more and more age-related diseases, we are
convinced that cell based therapies will play a
fundamental role in a continously increasing number of
indications. This has the potential to trigger the

30
development of an entire new generation of medicine for
the benefit of patients and of healthcare systems.
FIGURE 33  Examples of successful application of unmodified
reducing scar tissue. (a,b) Male, 48-year-old patient; scar tissue
formation on the upper arm after an injury caused by a car
accident; single application of UA-ADRCs isolated from 100 ml
lipoaspirate. (c,d) Male, 50-years-old patient; scar tissue
formation (arrows) in the face after an acid attack; three
treatments with UA-ADRCs isolated from 100 ml lipoaspirate on
days 1, 90 and 180. Abbreviations: BL, baseline; M1/M12,
one/twelve months after application of UA-ADRCs.
Ethics
The different studies during which cells were collected for
performing the experiments shown in Figs 2-6, 9-20 and
22-23 were approved by the Institutional Review Boards
(IRB) for Baylor College of Medicine and Affiliated
Hospitals (Houston, TX, USA) (Protocol # H-18357) and
University of Texas M. D. Anderson Cancer Center
(Houston, TX, USA) (Protocol # 03-08-03031 and 03-08-
03031).
Killing mice by cervical dislocation for isolating cells
from subcutaneous fat tissue (Muehlberg et al., 2009),
killing neonatal rats by decapitation for isolating
cardiomyocytes (Sadat et al., 2007) and killing adult rats
by decapitation for isolating cells (Alt and Bai,
unpublished data) did not require IRB approval but was
performed following the guidelines of Veterinary
Medicine & Surgery at the University of Texas M. D.
Anderson Cancer Center and the U.S. National Institutes
of Health.
Hallo
FIGURE 34  Examples of successful application of unmodified
treating hair loss. (a,b) 46-year-old female patient; sparse hair;
single application of UA-ADRCs isolated from 100 ml
lipoaspirate. (c-f) 23-years-old female patient suffering from
atopic dermatitis and alopecia totalis for three years without
spontaneous improvement; three treatments with UA-ADRCs
isolated from 100 ml lipoaspirate each at baseline (c) and at
three and six month after the first treatment. Results are shown
at one (M1; d), three (M3; e) and seven (M7; f) month after the
first treatment.
References
Ader, M. and E.M. Tanaka. 2014. "Modeling human development in
3D culture." Current Opinion in Cell Biology 31, 23–28.
Al-Drees, M. A. et al. 2015. „Making Blood: The Haematopoietic
Niche Throughout Ontogeny.” Stem Cells International 2015,
571893.
Alexander, C. E. and S. C. Dulebohn. 2018. „ Lumbosacral Facet
Syndrome”, StatPearls [Internet], Treasure Island (FL): StatPearls
Publishing; 2018-. 2017 Nov 5. (available online at
https://www.ncbi.nlm.nih.gov/books/NBK441906/).
Alt, E. et al. 2011. „Fibroblasts Share Mesenchymal Phenotypes with
Stem Cells, but Lack their Differentiation and Colony-Forming
Potential”, Biology of the Cell 103, 197–208.
Alt, E. 2015. „Mesenchymal Stem Cell - What Is in the Name?”,
Modern Medicine 22, 193–196.
Altman, A. M. et al. 2009. " IFATS Collection: Human Adipose-
Derived Stem Cells Seeded on a Silk Fibroin-Chitosan Scaffold
Enhance Wound Repair in a Murine Soft Tissue Injury Model", Stem
Cells 27:250–258.
Altman, A. M. et al. 2010. "Human Tissue-Resident Stem Cells
Combined with Hyaluronic Acid Gel Provide Fibrovascular-
Integrated Soft-Tissue Augmentation in a Murine Photoaged Skin
Model", Plastic and Reconstructive Surgery 125, 1–11.
Altman, A. M. et al. "Adipose Tissue–Derived Stem Cells Enhance
Bioprosthetic Mesh Repair of Ventral Hernias", Plastic and
Reconstructive Surgery 126, 845–854.
Altman, A. M. et al. 2011. „Wound Microenvironment Sequesters
Adipose-Derived Stem Cells in a Murine Model of Reconstructive
Surgery in the Setting of Concurrent Distant Malignancy”, Plastic
and Reconstructive Surgery 127, 1467–1477.
Andriolo, L. et al. 2018. „Regenerative Therapies Increase

Fundamentals of Stem Cells

Recommended

Recommended

More Related Content

What's hot

What's hot (19)

Similar to Fundamentals of Stem Cells

Similar to Fundamentals of Stem Cells (20)

Recently uploaded

Recently uploaded (20)

Fundamentals of Stem Cells